Ligand Dynamics of tert-Butyl Isocyanide Oxido Complexes ofMolybdenum(IV)Jana Leppin, Christoph Forster, and Katja Heinze*

Institute of Inorganic Chemistry and Analytical Chemistry, Johannes Gutenberg-University of Mainz, Duesbergweg 10−14, 55128Mainz, Germany

*S Supporting Information

ABSTRACT: The six-coordinate molybdenum(IV) oxido isocya-nide complex 1 [Δ,Λ-OC-6-2-3-[MoO(Np∩Ni)2(CN

tBu)]; Np∩Ni

= 4-tert-butylphenyl(pyrrolato-2-ylmethylene)amine] is obtained indiastereomerically pure form in the solid state, as revealed by single-crystal X-ray diffraction. In solution, this stereoisomer equilibrateswith the Δ,Λ-OC-6-2-4 diastereomer 2 at ambient temperature. Thestereochemistry of both isomers has been elucidated by NMR, IR,and UV/vis spectroscopy in combination with density functionaltheory (DFT)/polarizable continuum model and time-dependentDFT calculations. The isomerization 1 → 2 is suggested to proceedvia a dissociative trigonal twist with dissociation of the iminenitrogen donor Ni of one chelate ligand (hemilabile ligand) ratherthan dissociation of the monodentate isocyanide ligand. Theisomerization barrier has been experimentally determined as 91 and 95 kJ mol−1 in tetrahydrofuran and toluene, respectively.

■ INTRODUCTION

Oxidomolybdenum complexes are known to catalyze oxygen-atom-transfer (OAT) reactions between closed-shell mole-cules.1 A tremendous amount of work has been devoted toelucidate the mechanism with respect to biomimetic and otheruseful OAT reactions. Especially, molybdenum oxotransferasemodel systems have been studied in considerable detail.2

Molybdenum-dependent carbon monoxide dehydrogenases(CODHs) catalyze the reaction CO + H2O → CO2 + 2e− +2H+. The enzyme contains a unique molybdopterin−molybdenum active site in combination with a copper centerconnected by a sulfido bridge (Scheme 1).3 In the n-butylisocyanide-inhibited form of CODH, a bridging thio-carbamate is formed with N-coordinated copper (Scheme 1).3

A seminal structural model for molybdenum-dependentCODH with a MoVO−μ-S−CuI core has been reported byYoung et al.4 Hofmann et al. suggested that CO2 eliminationoccurs from molybdenum-bound κ2-(O,C)-CO2 (Scheme 1),while according to Siegbahn and Shestakov, CO2 is releasedfrom the copper center (Scheme 1).5,6 In any case, the catalyticcycle is completed by double deprotonation and doubleoxidation of the molybdenum(IV) intermediate.Molecular precedence for MoO nucleophilic attack at

[LnMCO]+, giving μ2-η3-CO2 ligands, has been reported by

employing electron-rich 18 valence-electron MoIVCp2(O)complexes and cationic carbonyl complexes.7 Recently, a ratherelectrophilic MoIVCp*(amidato)(O) complex was found tooxygenate CNtBu to OCNtBu via a MoIICp*(amidato)(κ2-(O,C)-OCNtBu)(CNtBu) intermediate in the presence ofexcess CNtBu.8 However, MoII oxidation states have been

neither observed nor postulated in molybdenum-containingoxygenases.Thus, we aimed at preparing analogue systems that bind a

carbon-based substrate at molybdenum in biologically relevant

Received: October 4, 2013Published: January 6, 2014

Scheme 1. (Top) Suggested Intermediates by Hofmann etal.5 and Siegbahn and Shestakov6 (S∩S = molybdopterin; X= O, NR) of Molybdenum-Dependent CODH and (Bottom)OAT between Molybdenum(IV) Oxido andMolybdenum(IV) Complexes with the Chelate LigandNp∩Ni

Article

pubs.acs.org/IC

© 2014 American Chemical Society 1039 dx.doi.org/10.1021/ic4025102 | Inorg. Chem. 2014, 53, 1039−1047

oxidation states below VI+, i.e., V+ and IV+. Recently, we haddescribed a molybdenum(VI,V,IV) system with unsymmetricalSchiff base ligands 4-tert-butylphenyl(pyrrolato-2-ylmethylene)-amine (Np∩Ni, where Ni = imine nitrogen atom and Np =pyrrolato nitrogen atom; Scheme 1) capable of catalyticallytransferring oxygen atoms to phosphanes as substrates fromdimethyl sulfoxide or from a water/ferrocenium oxidant asterminal oxygen donors.9−11 As a quite stable and fullycharacterized intermediate in the catalytic cycle, we isolatedthe MoIVO(Np∩Ni)2(PMe3) complex with a further substrate,stabilizing the MoIV core. Similarly, the analogous imidocomplex MoIV(NtBu)(Np∩Ni)2(PMe3) and its one-electron-oxidized redox congener [MoV(NtBu)(Np∩Ni)2(PMe3)]

+ havebeen isolated and fully characterized by NMR and electronparamagnetic resonance spectroscopy, density functional theory(DFT) calculations, and X-ray diffraction (Scheme 1).12 Wethus speculated that tert-butyl isocyanide as a CO surrogatecould be coordinated to MoIV (and possibly also to MoV) inplace of PMe3 in a similar ligand environment, leading toprecedence of truncated (copper-deficient) potential inter-mediates MoO(Np∩Ni)2(CN

tBu) or Mo(Np∩Ni)2(OCNtBu).

With this background, we investigated the preparation,isolation, and stability/lability of molybdenum oxido isocyanidecomplexes based on the chelate ligand system Np∩Ni.Typical experimental access pathways to MoIVO(CNtBu)

complexes involve CNtBu substitution in trans-MoIV(Cl)O-(CNtBu)4

13,14 by chelating ligands,15,16 N2 substitution inMoIII(N2) complexes by CNtBu and subsequent oxidation toMoIV17 and ligand exchange at molybdenum(IV) phosphanecomplexes.18,19 Oxidative addition of OCNtBu has beenreported to be viable for tungsten(II) complexes possibly viaa κ2-(O,C)-OCNtBu intermediate,20 while oxidative addition ofOCNtBu to MoII only yields a MoIV(NtBu)(CO) complex.21

■ EXPERIMENTAL SECTIONGeneral Procedures. All reactions were performed under an inert

atmosphere (Schlenk techniques, glovebox). Tetrahydrofuran (THF)was distilled from potassium, diethyl ether, and petroleum ether (bp40−60 °C) and triethylamine from calcium hydride. The Schiff baseligand HNp∩Ni and the phosphane complex MoIVO(Np∩Ni)2(PMe3)were prepared according to literature procedures.12 All other reagentswere used as received from commercial suppliers (Acros, Sigma-Aldrich). NMR spectra were recorded on a Bruker Avance DRX 400spectrometer at 400.31 MHz (1H), 100.66 MHz (13C{1H}), and 40.56MHz (15N). All resonances are reported in ppm versus the solventsignal as the internal standard [THF-d8 (

1H, δ 1.73, 3.58; 13C, δ 25.37,67.57)] versus external CH3NO2 [90% in CDCl3 (

15N, δ 380.23)]. 15Ndata are reported versus liquid NH3 as the reference (δ 0). IR spectrawere recorded with a BioRad Excalibur FTS 3100 spectrometer as CsIdisks or in KBr cells in solution. UV/vis/near-IR spectra wererecorded on a Varian Cary 5000 spectrometer using 1.0 cm cells(Hellma, Suprasil). Field-desorption (FD) mass spectra were recordedon a FD Finnigan MAT95 spectrometer. Elemental analyses wereperformed by the microanalytical laboratory of the Chemical Institutesof the University of Mainz.Crystal Structure Determination. Intensity data were collected

with a Bruker AXS Smart 1000 CCD diffractometer with an APEX IIdetector and an Oxford cooling system and corrected for absorptionand other effects using Mo Kα radiation (λ = 0.71073 Å) at 173(2) K.The diffraction frames were integrated using the SAINT package, andmost were corrected for absorption with SADABS.22,23 The structurewas solved by direct methods and refined by the full-matrix methodbased on F2 using the SHELXTL software package.24,25 All non-hydrogen atoms were refined anisotropically, while the positions of allhydrogen atoms were generated with appropriate geometricconstraints and allowed to ride on their respective parent carbon

atoms with fixed isotropic thermal parameters. The THF solvatemolecule shows pseudorotational disorder with occupancy factors of0.3 and 0.7. Crystallographic data (excluding structure factors) for thestructure reported in this paper have been deposited with theCambridge Crystallographic Data Centre as supplementary publicationCCDC 952600. Copies of the data can be obtained free of chargeupon application to CCDC, 12 Union Road, Cambridge CB2 1EZ,U.K. [fax (0.44) 1223-336-033; e-mail deposit@ccdc.cam.ac.uk].

Crystallographic data of 1: C39H51MoN5O2 (717.79); ortho-rhombic; Pbca; a = 17.7739(13) Å, b = 16.1821(9) Å, c = 25.8978(16)Å, V = 7448.7(8) Å3; Z = 8; density, calcd = 1.280 g cm−3, μ = 0.391mm−1; F(000) = 3024; crystal size 0.50 × 0.35 × 0.20 mm; θ = 1.57−28.02°; −23 ≤ h ≤ 23, −21 ≤ k ≤ 17, −34 ≤ l ≤ 34; reflns collected =49312; reflns unique = 8995 [R(int) = 0.0479]; completeness to θ =28.02° = 99.7%; semiempirical absorption correction from equivalents;max and min transmission = 0.926 and 0.828; no. of data = 8995; no.of restraints = 0; no. of parameters = 442; GOF on F2 = 1.138; final Rindices [I > 2σ(I)] of R1 = 0.0323, wR2 = 0.0763; R indices (all data)of R1 = 0.0552, wR2 = 0.0909; largest difference peak and hole =0.339 and −0.357 e Å−3.DFT calculations were carried out with the Gaussian09/DFT26 seriesof programs. The B3LYP formulation of DFT was used by employingthe LANL2DZ basis set supplemented by d-type polarizationfunctions27 on nitrogen (ζ = 0.864) and oxygen (ζ = 1.154). Allstructures were characterized as minima or first-order saddle points byfrequency analysis (Nimag = 0, 1). No symmetry constraints wereimposed on the molecules. Solvent modeling was done by employingthe integral equation formalism polarizable continuum model(IEFPCM, THF). The modeled complexes were slightly simplifiedby replacing the tert-butyl group of the chelate ligands by hydrogenatoms. The approximate free energies at 298 K were obtained throughthermochemical analysis of the frequency calculation, using thethermal correction to the Gibbs free energy, as reported byGaussian09.

Synthesis of 1/2.MoIVO(Np∩Ni)2(PMe3)12 (350 mg, 0.55 mmol)

was dissolved in THF (6 mL), and tert-butyl isocyanide (620 μL, 5.5mmol) was added. The mixture was stirred for 24 h at 50 °C under aslow flow of argon, removing the volatile PMe3. After cooling to roomtemperature, all volatiles were removed under reduced pressure to givethe 1/2 mixture as a green powder in essentially quantitative yield.

Isolation of 1 from the 1/2 Mixture. By recrystallization of the1/2 mixture from petroleum ether (bp 40−60 °C) at 8 °C,isomerically pure crystalline 1 was isolated in 21% yield (74 mg,0.12 mmol). Elem anal. Calcd for C35H43N5OMo (645.70)·1/2hexane:C, 66.26; H, 7.32; N, 10.17. Found: C, 66.87; H, 7.54; N, 10.67. FD-MS: m/z (%) = 647.4 (22) ([M]+; correct isotopic distribution), 564.4(9) ([M − CNtBu]+). IR (CsI): ν 3092 (w, CHar), 2962 (m, CHal),2129 (m, CN), 1568 (vs), 1508 (m), 1392 (m), 1291 (s), 1179(m), 1036 (s), 937 (vs, MoO), 756 (vs), 746 (vs) cm−1. UV/vis[THF; λmax, nm (ε, M−1 cm−1)]: 255 (22605, sh), 307 (31265), 402(12355), 540 (2440). 1H NMR (THF-d8): δ 8.13 (s, 1H, H

7b), 7.69 (s,1H, H11b), 7.44 (s, 1H, H7a), 7.24 (d, 3JHH = 8.7 Hz, 2H, H3a,5a), 7.15(d, 3JHH = 8.7 Hz, 2H, H3b,5b), 7.06 (d, 3JHH = 8.7 Hz, 2H, H2a,6a), 7.01(dvd, 1H, H9b), 6.77 (d, 3JHH = 8.7 Hz, 2H, H2b,6b), 6.47 (dvd, 1H,H10b), 6.28 (dvd, 1H, H9a or H10a), 5.75 (m, 1H, H11a), 5.73 (dvd, 1H,H9a or H10a), 1.44 (s, 9H, H16a,b), 1.35 (2 s, 18H, H13a,b). 13C{1H}NMR (THF-d8): δ 158.9 (s, C

7b), 153.0 (s, C7a), 152.4 (s, C1a), 150.5(s, C1b), 148.7 (s, C4a), 149.2 (s, C4b), 146.9 (s, C11b), 144.4 (s, C8b),140.9 (s, C8a), 137.9 (s, C11a), 126.3 (s, C3b/5b), 125.7 (s, C3a/5a), 124.3(s, C2a/6a), 123.1 (s, C2b/6b), 119.6 (s, C9b), 116.1 (s, C9a or C10a),115.1 (s, C10b), 113.6 (s, C9a or C10a), 60.6 (s, CNC(CH3)3), 35.0(s, C12a,12b), 31.9 (s, C13a,13b), 31.0 (s, CNC(CH3)3); the resonanceof CNC(CH3)3 is not observed.

15N NMR (THF-d8): δ 274.9 (NC), 238.2 (Ni,b), 232.7 (Np,a), 230.2 (Ni,a), 220.2 (Np,b).

Synthesis of the 1/2 Isomer Mixture. A THF solution of 1 wasallowed to equilibrate at room temperature for at least 2 h, and thespectroscopic data for 2 were determined from this equilibriummixture. 1H NMR (THF-d8): δ 8.41 (s, 1H, H7d), 7.99 (s, 1H, H7c),7.74 (d, 3JHH = 8.7 Hz, 2H, H2c,6c), 7.60 (d, 3JHH = 8.7 Hz, 2H, H2d,6d),7.52 (d, 3JHH = 8.7 Hz, 2H, H3d,5d), 7.44 (d, 3JHH = 8.7 Hz, 2H, H3c,5c),

Inorganic Chemistry Article

dx.doi.org/10.1021/ic4025102 | Inorg. Chem. 2014, 53, 1039−10471040

7.34 (m, 1H, H11d), 6.98 (dvd, 1H, H9d), 6.42 (m, 1H, H11c), 6.35(dvd, 1H, H9c), 6.21 (dvd, 1H, H10d), 5.78 (dvd, 1H, H10c), 1.38 (2 s,18H, H13c,13d) 1.01 (s, 9H, H16c,16d). 13C{1H} NMR (THF-d8): δ 157.8(s, C7d), 151.4 (s, C7c), 154.1 (s, C1d), 151.1 (s, C1c), 149.2 (2 s,C4c,4d), 143.3 (s, C8d), 141.4 (s, C8c), 141.5 (s, C11d), 138.1 (s, C11c),126.6 (s, C3d/5d), 126.5 (s, C3c/5c), 123.4 (s, C2c/6c), 123.3 (s, C2d/6d),121.6 (s, C9d), 117.5 (s, C9c), 115.0 (s, C10d), 113.6 (s, C10c), 58.7 (s,CNC(CH3)3), 35.2 (s, C12c,12d), 31.9 (s, C13c,13d), 30.2 (s, CNC(CH3)3); the resonance of CNC(CH3)3 is not observed. 15NNMR (THF-d8): δ 250.5 (NC), 231.5 (Np,d), 228.6 (Np,c), 225.2(Ni,c), 221.3 (Ni,d). UV/vis [λmax, nm (ε, M−1 cm−1)]: 235 (19765),307 (35700), 449 (9970), 537 (7705) (data obtained from thedifference spectrum “1/2 mixture minus 1”).Reaction of A with Excess CNtBu. The dioxido complex A (20

mg, 0.034 mmol) was dissolved in toluene (4 mL), and CNtBu (14.3mg, 0.17 mmol) was added. The solution was heated to 95 °C forseveral hours, and 1H NMR spectra were measured. After 5.5 h, theonly resonances observed are those of A, CNtBu, and free chelateligand.Reaction of 1/2 with Water/Base. The isomer mixture 1/2 (13.2

mg, 0.02 mmol) was dissolved in THF-d8 (1 mL). Phosphazene basetert-butyliminotris(dimethylamino)phosphorane (P1

tBu; 9.56 mg, 0.04mmol) and water (0.74 μL, 0.04 mmol) were added. According to 1HNMR spectra, the mixture remained unchanged for several weeks insolution at room temperature, except for some chelate liganddissociation.

■ RESULTS AND DISCUSSIONSynthesis. The previously reported molybdenum(VI)

complex MoVIO2(Np∩Ni)2 (A) is competent to oxygenate

phosphanes, e.g., PMe3, giving the molybdenum(IV) phos-phane complex MoIVO(Np∩Ni)2(PMe3) (B) and phosphaneoxide.9−12 In an analogous experiment, A is treated with excesstert-butyl isocyanide, but OAT to CNtBu is not observed eitherat room temperature or at elevated temperature (95 °C,toluene).Coordination of CNtBu at the molybdenum(IV) center to

give MoIVO(Np∩Ni)2(CNtBu) was achieved by PMe3 sub-

stitution18,19 from molybdenum(IV) complex B and perma-nently removing the liberated volatile PMe3 by a slow flow ofargon to drive the reaction to completion (Scheme 2).According to IR and NMR spectroscopy (vide infra), no attackat the oxido ligand of the MoIVO unit to give coordinated (or

liberated) OCNtBu and MoII was observed, as has beendemonstrated for a MoIVCp*(amidato)(O) complex andCNtBu, giving a κ2-(C,O)-OCNtBu-coordinated molybdenum-(II) complex.8

Stereochemistry. After recrystallization of the resultinggreen material with the composition MoIVO(Np∩Ni)2(CN

tBu)(by mass spectrometry) from 1:2 THF/petroleum ether (bp40−60 °C), the 1H NMR spectrum of the green compound,taken immediately after dissolution, features resonancescorresponding to two chelate ligands Np∩Ni and onecoordinated isocyanide ligand (δ 1.44) in the expected intensityratio (Figure 1, bottom). Upon standing at room temperature,these resonances lose intensity. Concomitantly, new resonancescorresponding to two further chelate ligands (14 1Hresonances) and one isocyanide ligand (δ 1.01) develop(Figure 1). A stationary state is reached after approximately 2h (≈2:1 ratio). At higher temperature, the equilibriumcomposition does not change appreciably, suggesting onlysmall entropy changes during the reaction. Resonancescorresponding to four chelate ligands Np∩Ni and twoisocyanides in total are also detected in the 13C{1H} NMRand the 1H15N HMBC spectra (4 × Np, 4 × Ni, and 2 ×Nisocyanide; Figures S1 and S2 in the SI). These observationsclearly point to an equilibrium between two stereoisomers insolution (Scheme 2 and Chart 1). As can be seen from Figure 1,tiny amounts of other diastereomers are also formed.In order to assign the configuration of the isomers formed,

nuclear Overhauser effect (NOE) spectral data were acquiredand analyzed with the aid of DFT models (Figure 2 and Table1; B3LYP/LANL2DZ/PCM). Short distances between theisocyanide protons and the chelate ligand orthogonal to theMo−C vector (ligands a, c, e, and g in 1−4) are possible in allstereoisomers and are thus inconclusive for assignments of thestereochemistry (Figure 2). However, short distances betweenthe isocyanide protons and the second chelate (ligands b, d, f,and h in 1−4) discriminate between pairs of stereoisomers 1/3and 2/4. Stereoisomers 1 and 3 show a short contact to H11

(pyrrolate of ligands b and f coplanar to isocyanide Mo−CN; d≤ 3 Å), while diasteromers 2 and 4 feature short contacts toH2/6 and H3/5 (NiAr of ligands d and h coplanar to isocyanideMo−CN; d ≤ 2.2 Å). The major isomer present at thebeginning of the isomerization directly after dissolution displaysfive NOE cross peaks between the isocyanide protons (δ 1.44)and one chelate ligand, as expected, and a further cross peak toH11 of the second chelate, which is hence compatible withconfigurations 1 and 3, respectively (Figure 2 and Chart 1).The second (minor) isomer progressively formed at T > −30°C gives three NOE cross peaks between the isocyanideprotons (δ 1.01) and the chelate ligands. Contacts areestablished between the isocyanide and the pyrrole protonH11 of one chelate and the aromatic protons H2,6 and H3,5 ofthe other chelate. These contacts are compatible withconfigurations 2 and 4 (Chart 1).Fortunately, single crystals of the isocyanide complex were

obtained from a 1:2 THF/petroleum ether (bp 40−60 °C)solution at 8 °C. X-ray diffraction analysis (orthorhombic Pbca,THF solvate) substantiates configuration 1 (OC-6-2-3) in thecrystalline state (Figure 3). The experimental metrical data ofthe complex also fit to the DFT-calculated data of stereoisomer1 (Table 1). The MoO distance of 1.685 Å is similar to thatof the PMe3 complex B [MoIVO(Np∩Ni)2(PMe3) withanalogous stereochemistry: d(Mo−O) = 1.686 Å]12 and tothat of a molybdenum(IV) oxido bis(isocyanide) complex with

Scheme 2. Synthesis of Molybdenum(IV) Oxido tert-ButylIsocyanide Complexes from the Dioxido Complex A via thePhosphane Complex B

Inorganic Chemistry Article

dx.doi.org/10.1021/ic4025102 | Inorg. Chem. 2014, 53, 1039−10471041

a tris(mercaptoimidazolyl)borate ligand [d(Mo−O) = 1.684Å].15 The Mo−C bond lengths of the latter complex are smaller[d(Mo−C) = 2.021/2.075 Å] than that of 1 [d(Mo−C) =2.109 Å]. No O···C interaction is observed in 1 [d(O···C) =2.785 Å], as found in the complex MoCp*(amidato)(κ2-(O,C)-OCNtBu)(CNtBu) [d(O···C) = 1.307 Å] with side-on κ2-(O,C)-OCNtBu coordination.8

Having assigned the OC-6-2-3 configuration to isomer 1present in the solid state, we attempted to elucidate theconfiguration of the second isomer (2 or 4 according to NOEdata). Solid-state IR spectra recorded for the recrystallizedmaterial (pure isomer 1) feature an absorption band for the CNstretching mode at νCN = 2129 cm−1, while the precipitated rawmaterial of the isomer mixture shows a broad absorption bandat νCN = 2142 cm−1 (Figure S3 in the SI). DFT calculations andharmonic frequency calculations yield CN stretching vibrationsfor the stereoisomers 1−4 at νCN = 2117, 2124, 2113, and 2113cm−1, respectively (Table 1, scaled by 0.961433). These datasupport configuration 1 for the major isomer and suggestconfiguration 2 for the minor isomer with a slightly higher CNstretching frequency than 1. Because both configurations 3 and4 possess lower CN stretching frequencies than 1 according tothe calculations, they are likely excluded as minor isomers.In addition, the MoO stretching vibrations are calculated

with very similar energies for the pairs 1/2 (anionic pyrrolatetrans to oxido; 947/946 cm−1) and 3/4 (π-accepting imine

trans to oxido; 965/967 cm−1). The recrystallized materialdisplays νMoO = 937 cm−1, similar to the raw material (νMoO =939 cm−1; Figure S3 in the SI). Because the MoO stretchingmode is essentially insensitive to the nature of the cis ligandsbut very sensitive to the trans ligand,14 the νMoO data suggestthat the trans ligand remains at its position during isomer-ization. Thus, according to the MoO stretching mode, 1 istransformed to 2 or 3 to 4, with the latter case being excludedon the basis of the CN stretch and X-ray diffraction. Thus, allavailable data point to the isomerization of 1 to 2 beingoperative with only minor isomerization to isomers 3 or 4.Gratifyingly, the two isomers 1 and 2 are also calculated as themost stable ones by DFT calculations (Table 1). In a THFsolution, the CN stretching vibration shifts from νCN = 2135cm−1 (1) to 2138 cm−1 (mixture of 1 and 2) within 2 h at roomtemperature.Similar to evolution of the NMR and IR spectra, the optical

properties change upon dissolution of crystals of 1 in THF untilequilibrium is reached (Figure 4). From the NMR data, thefinal ratio (≈2:1 at 298 K) is known. With this ratio, theapproximate spectrum of the minor isomer can be extractedfrom the final UV/vis spectrum (Figure 5, top; note that otherisomers formed in tiny amounts are neglected). The π−π*absorptions of the minor isomer are more intense than thosefound for 1 (≈1:1.14) but remain essentially unshifted (λmax =307 nm), and an absorption window (≈390−400 nm) arisesbetween the π−π* and ligand field/CT bands in the minorisomer.In order to identify significant optical differences among

isomers 1−4, TD-DFT calculations have been performed(Figure 5, bottom). Isomers 1 and 2 feature π−π* absorptionscalculated at similar energy [λmax(1) = λmax(2) = 307 nm] witha higher intensity for 2 (1:2 ≈ 1:1.35). On the other hand,isomers 3 and 4 are calculated to possess bathochromicallyshifted π−π* bands [λmax(3) = 332 nm and λmax(4) = 326 nm]with lower intensity compared to 1 (1:3 ≈ 1:0.86 and 1:4 ≈1:0.87). The data pertaining to 3 and 4 are thus incompatiblewith the experimental findings (Figure 5). Furthermore, isomer1 features a CT absorption band at 402 nm (Figure 5, top),

Figure 1. 1H NMR spectra of MoIVO(Np∩Ni)2(CNtBu) in THF-d8 directly after dissolution of crystals (bottom) and after 30 min (top) at 25 °C.

Growing resonances are marked with asterisks and arrows.

Chart 1. Possible Diastereomers 1−4, StereodescriptorsBased on the CIP Priorities28−32 Shown in Red, andIndicated Chelate Ligands a−h (Corresponding ΔEnantiomers Not Shown)

Inorganic Chemistry Article

dx.doi.org/10.1021/ic4025102 | Inorg. Chem. 2014, 53, 1039−10471042

which is calculated around 355 nm by TD-DFT (Figure 5,bottom). The transitions of this band at 347/364 nm ( f =0.055/0.050) are basically pyrrolate to π*(MoO/imine) incharacter. For isomer 2, the energies of these transitions aremore distinct and the intensities are much weaker (354/380nm, f = 0.014/0.014), resulting in less absorption in the spectralregion of the CT transitions, in agreement with the experiment(Figure 5). On the basis of the combined NMR, IR, and UV/visdata, we assign configuration 2 (OC-6-2-4) to the minor isomerwith high confidence.Kinetics of Isomerization. In general, the isomerization of

1 to 2 (and to other diastereomers and enantiomers) couldproceed via a trigonal-twist mechanism or via a dissociativemechanism with a genuine five-coordinate intermediate. In thetwist mechanism, rearrangement occurs via rotation about apseudo-3-fold axis of the idealized coordination octahedron. Apseudorotation mechanism has been suggested to be operative

in ReVO(dithiolate)(CH3)(L) complexes34 and cis/trans-MoVO(NNO)Cl2 as well as MoVIO2(NNO)Cl complexes(NNO = heteroscorpionato ligand).35 The dissociativemechanism proceeds via ligand loss, subsequent Berry orturnstile pseudorotation of the five-coordinate intermediate,and recoordination at a stereochemically different site. Adissociative mechanism has been invoked in cis/trans-ReVO-(CH3)2Cl(bpy) via a monodentate 2,2′-bipyridine (bpy)ligand.36 Hemilability of a bpy chelate has also been suggestedfor FeEt2(bpy)2.

37 Potentially labile Mo−ligand bonds in 1 arecertainly the Mo−C or Mo−Ni bonds, while the MoOmultiple bond and the Mo−Np bonds are considered to bemuch stronger. The possibility of Mo−C cleavage is supportedby the mass spectrum of 1 showing M+ and [M − CNtBu]+

peaks, albeit for the cation. Naturally, intermediate partialdissociation of a chelate Ni donor would be invisible to massspectrometric analysis.

Figure 2. DFT-calculated minimum geometries of stereoisomers 1−4 and relevant interligand H···H distances (minimum distances/Å achievable byrotation and torsion around single bonds).

Inorganic Chemistry Article

dx.doi.org/10.1021/ic4025102 | Inorg. Chem. 2014, 53, 1039−10471043

Isocyanide dissociation has been calculated by DFT methodsfrom 1 and 2 to give the five-coordinate cis and trans isomers 5,

respectively (Chart 2). The free energies of activation arecalculated as 83 and 63 kJ mol−1 for TS(Λ-1→cis-5) and TS(Λ-2→trans-5), respectively (Figure 6), which are reasonablevalues for CNtBu dissociation from MoIV. The necessary five-

Table 1. Experimental and DFT-Calculateda Relevant Metrical (Å and deg), Vibrational (cm−1), and Energy Data (kJ mol−1) of1−4

X-ray 1 (x = a; y = b) DFT 1 (x = a; y = b) DFT 2 (x = c; y = d) DFT 3 (x = e; y = f) DFT 4 (x = g; y = h)

Mo1−O1 1.6850(15) 1.701 (1.708) 1.703 (1.709) 1.691 (1.699) 1.690 (1.698)Mo1−C1 2.1093(22) 2.108 (2.120) 2.126 (2.136) 2.098 (2.110) 2.114 (2.127)Mo1−Nx,i 2.2135(17) 2.285 (2.285) 2.229 (2.231) 2.493 (2.475) 2.507 (2.493)Mo1−Nx,p 2.2422(17) 2.273 (2.268) 2.312 (2.306) 2.186 (2.193) 2.165 (2.170)Mo1−Ny,i 2.1980(17) 2.252 (2.254) 2.202 (2.203) 2.236 (2.236) 2.229 (2.227)Mo1−Ny,p 2.1074(18) 2.119 (2.116) 2.142 (2.141) 2.136 (2.131) 2.134 (2.131)C1−N1 1.1552(27) 1.177 (1.174) 1.177 (1.175) 1.178 (1.175) 1.178 (1.175)Mo1−C1−N1 176.25(19) 179.3 (179.8) 177.6 (178.4) 179.7 (179.2) 176.6 (176.1)C1−Mo1−O1 93.76(8) 94.1 (94.2) 90.8 (90.7) 95.7 (95.6) 92.5 (92.6)ν(MoO) 937 947 (927)b 946 (928)b 965 (942)b 967 (945)b

ν(CN) 2129 2117 (2129)b 2124 (2130)b 2113 (2122)b 2113 (2121)b

ΔG°(298 K) 0.0 (0.0) 1.1 (2.5) 5.6 (5.8) 7.4 (9.5)aFrom model complexes with the tert-butyl group of the chelate ligands replaced by hydrogen atoms; gas-phase calculations; PCM(THF) data inparentheses. bCalculated harmonic frequencies are scaled by 0.9614.33

Figure 3. (a) Molecular structure of 1 in the crystal (hydrogen atomsomitted for clarity; thermal ellipsoids with 50% probability) and (b)photograph of crystals of 1.

Figure 4. UV/vis spectra of MoIVO(Np∩Ni)2(CNtBu) in THF after

dissolution of crystals of 1 at −30 °C and warming to +25 °C.

Figure 5. UV/vis spectra of 1 in THF (black ) and of 2 (obtainedfrom the difference spectrum “1/2 mixture minus 1”) in THF (red- - -) [top] and those from TD-DFT calculations of 1 (black ), 2(red - - -), 3 (green -·-), and 4 (blue ···) and Lorentzian band shapeswith fwhm 25 cm−1 [bottom].

Inorganic Chemistry Article

dx.doi.org/10.1021/ic4025102 | Inorg. Chem. 2014, 53, 1039−10471044

coordinate turnstile transition states for isomerization from cis-5 to trans-5 are TS1(cis-5→trans-5) and TS2(cis-5→trans-5)with Np or Ni in pseudoapical positions (Figure 6). Thesetransition states are calculated with barriers of 149/148 kJmol−1, which seem to be quite high. The calculations suggestthat the Mo−Ni bond of one chelate is elongated duringpseudorotation. In summary, the turnstile rotation requiresmuch more energy than CNtBu dissociation and becomes therate-determining step (rds) in this mechanism. For complete-ness, the barriers for CNtBu dissociation from 3 to trans-5 and4 to cis-5 have also been calculated as 90 and 76 kJ mol−1 (seethe SI).Pseudorotation of the trigonal face spanned by O, C, and one

Ni against the opposite trigonal plane (trigonal-twist mecha-nism) gives isomer Δ-2 from Λ-1 at much lower energy cost(65 kJ mol−1; Figure 7). Inspection of the pseudo-trigonal-prismatic transition-state structure TS(Λ-1→Δ-2) reveals thatduring the twist one Mo−Ni bond is completely disrupted.Hence, the mechanism of isomerization is best described by adissociative pseudorotation involving a Ni donor dissociation ofone chelate ligand. This interpretation is in accordance with the

mechanism proposed for the cis/trans isomerization ofReO(CH3)2Cl(bpy) on the basis of its small positive activationentropy and rate independence of added bpy.36 Because in bothcalculated scenarios the rds, namely, either TS(cis-5→trans-5)or TS(Λ-1→Δ-2), is unimolecular, the equilibration rateshould not be diminished by the presence of excess CNtBuin either case. Indeed, in the presence of 2, 10, or 100 equiv ofCNtBu, the reaction is not slowed. The reaction proceeds evenfaster according to 1H NMR measurements. 50% conversion atroom temperature is achieved at 895, 350, 155 and <100 s for 0,2, 10, and 100 equiv of CNtBu added. This points to thepossibility of an additional associative (or interchange Ia)

Chart 2. Five-Coordinate Complexes cis- and trans-5

Figure 6. Calculated energy profile [ΔG°(298 K)/kJ mol−1] of the isomerization of Λ-1 to -2 via cis- and trans-5 intermediates (B3LYP/LANL2DZ/PCM THF; hydrogen atoms omitted for clarity).

Figure 7. Calculated energy profile [ΔG°(298 K)/kJ mol−1] of theisomerization of Λ-1 to Δ-2 via a dissociative twist (B3LYP/LANL2DZ/PCM THF; hydrogen atoms omitted for clarity).

Inorganic Chemistry Article

dx.doi.org/10.1021/ic4025102 | Inorg. Chem. 2014, 53, 1039−10471045

pathway in the presence of excess CNtBu. In the absence ofexcess CNtBu, this pathway seems to play only a minor role.With the condition that the rds in the isomerization

equilibrium 1 to 2 (without added CNtBu) is first order in 1,we studied the process by 1H NMR spectroscopy at differenttemperatures and analyzed the reaction progress according to afirst-order process. From time-dependent 1H NMR measure-ments performed between 288 and 318 K, time traces of theevolution of 1 to 2 were obtained from integration of H7

proton resonances at δ 8.41 (H7d) of isomer 2 or isocyanideproton resonances of 2 at δ 1.01. Other isomers 3 and 4 wereneglected. No accumulation of free isocyanide is observedduring equilibration. The experimental data have been fitted tothe rate law of a reversible first-order reaction d[2]/dt = kf[1]− kb[2]

= + − − += k k k k k k t2 1[ ] {[ ] /( )}{ exp[ ( ) ]}t 0 f b f f f b

and rate constants kf at different temperatures have beenextracted. Eyring−Polanyi plots give ΔH⧧

f = 91 and 95 kJmol−1 in THF-d8 and toluene-d8, respectively (Figure S4 in theSI). Because of the large errors associated with determination ofthe activation entropies using this small temperature range, werefrain from discussing ΔS⧧ values. However, the similarity ofthe activation enthalpies in THF and toluene suggests that thesolvent plays only a minor role in the rds, and an associative rdsinvolving THF is safely excluded.38 However, the experimentaldata do not distinguish between the two rather dissociative first-order mechanisms (Figures 6 and 7).Hence, the molybdenum(IV) oxido isocyanide complexes 1

and 2 are stereochemically flexible but appear stable toward(irreversible) ligand loss. 1 and 2 are even stable in thepresence of hydroxide (phosphazene base P1

tBu and H2O), andattack of hydroxide at either molybdenum or the isocyanideligand is not detected by NMR. One-electron-oxidationreactions of these stable molybdenum(IV) complexes 1 and 2to molybdenum(V) and their much more complex reactivitywill be reported in a separate contribution.

■ CONCLUSION

In this study, we aimed at obtaining a simple (copper-deficient)CODH analogue system with a carbon-based substrate(CNtBu) bound to molybdenum in the biologically relevantoxidation state IV+. OAT from MoO2(N

p∩Ni)2 to CNtBu wasnot observed; however, PMe3 ligand displacement fromMoO(Np∩Ni)2(PMe3) was feasible to give stereoisomericallypure molybdenum(IV) oxido isocyanide complex 1 afterrecrystallization. Its solid-state structure has been determinedby X-ray diffraction. The CNtBu ligand is coordinated in anend-on fashion to MoIV rather than in a η2-OCNtBu fashion, assuggested in the Hofmann intermediate (Scheme 1). However,the stereoisomer 1 equilibrates with the Δ,Λ-OC-6-2-4stereoisomer 2 in solution at ambient temperature. Theconfigurations of both diastereomers have been elucidated byNMR, IR, and UV/vis spectroscopy in combination with DFT/PCM and TD-DFT calculations. Mechanistically, the isomer-ization 1 → 2 is suggested to proceed via a dissociative trigonaltwist with dissociation of the imine nitrogen donor Ni of onechelate ligand (hemilabile ligand) rather than dissociation ofthe monodentate isocyanide ligand. The isomerization barrierhas been determined as 91 and 95 kJ mol−1 in THF andtoluene, respectively. The stability of the MoIV−C bond in 1and 2 paves the way for future investigations with respect to

higher oxidation states of molybdenum and with relevance forbiomimetic oxygenation reactions of carbon-based substrates.

■ ASSOCIATED CONTENT*S Supporting InformationNMR spectra of 1 and the 1 + 2 mixture, IR spectra of 1 andthe 1 + 2 mixture, Eyring−Polanyi plots, and Cartesiancoordinates and energies of all optimized structures. Thismaterial is available free of charge via the Internet at http://pubs.acs.org.

■ AUTHOR INFORMATIONCorresponding Author*E-mail: katja.heinze@uni-mainz.de. Fax: +49-6131-39-27277.Author ContributionsThe manuscript was written through contributions of allauthors. All authors have given approval to the final version ofthe manuscript.NotesThe authors declare no competing financial interest.

■ ACKNOWLEDGMENTSWe are grateful to Dr. Mihail Mondeshki for excellent NMRsupport and to Regine Jung-Pothmann and Dr. DieterSchollmeyer for X-ray data collection.

■ REFERENCES(1) Holm, R. H. Chem. Rev. 1987, 87, 1401−1449.(2) (a) Dobbek, H.; Gremer, L.; Meyer, O.; Huber, R. In Handbookof Metalloproteins; Messerschmidt, A., Huber, R., Wieghardt, K.,Poulos, T., Eds.; Wiley: Chichester, U.K., 2001; Vol. 2, pp 1136−1147.(b) Enemark, J. H.; Cooney, J. J. A.; Wang, J.-J.; Holm, R. H. Chem.Rev. 2004, 104, 1175−1200. (c) Schulzke, C. Eur. J. Inorg. Chem. 2011,1189−1199. (d) Sugimoto, H.; Tsukube, H. Chem. Soc. Rev. 2008, 37,2609−2619.(3) Dobbek, H.; Gremer, L.; Kiefersauer, R.; Huber, R.; Meyer, O.Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 15971−15976.(4) Gourlay, C.; Nielsen, D. J.; White, J. M.; Knottenbelt, S. Z.; Kirk,M. L.; Young, C. G. J. Am. Chem. Soc. 2006, 128, 2164−2165.(5) Hofmann, M.; Kassube, J. K.; Graf, T. J. Biol. Inorg. Chem. 2005,10, 490−495.(6) Siegbahn, P. E. M.; Shestakov, A. F. J. Comput. Chem. 2005, 26,888−898.(7) Pilato, R. S.; Geoffroy, G. L.; Rheingold, A. L. J. Chem. Soc., Chem.Commun. 1989, 1287−1288.(8) Yonke, B. L.; Reeds, J. P.; Zavalij, P. Y.; Sita, L. R. Angew. Chem.2011, 123, 12550−12554; Angew. Chem., Int. Ed. 2011, 50, 12342−12346.(9) Heinze, K.; Fischer, A. Eur. J. Inorg. Chem. 2007, 1020−1026.(10) Heinze, K.; Marano, G.; Fischer, A. J. Inorg. Biochem. 2008, 102,1199−1211.(11) Heinze, K.; Fischer, A. Eur. J. Inorg. Chem. 2010, 1939−1947.(12) Huttinger, K.; Forster, C.; Bund, T.; Hinderberger, D.; Heinze,K. Inorg. Chem. 2012, 51, 4180−4192.(13) Novotny, M.; Lippard, S. J. Inorg. Chem. 1974, 13, 828−831.(14) Re, R. E. D.; Hopkins, M. D. Inorg. Chem. 2002, 41, 6973−6985.(15) Tran, B. L.; Carrano, C. J. Inorg. Chem. 2007, 46, 5429−5438.(16) Soncksen, L.; Romer, R.; Nather, C.; Peters, G.; Tuczek, F.Inorg. Chim. Acta 2011, 374, 472−479.(17) Greco, G. E.; O’Donoghue, M. B.; Seidel, S. W.; Davis, W. M.;Schrock, R. R. Organometallics 2000, 19, 1132−1149.(18) Carmona, E.; Galindo, A.; Guille-Photin, C.; Sanchez, L.Polyhedron 1988, 7, 1767−1771.(19) Montilla, F.; Galindo, A.; Carmona, E.; Gutierrez-Puebla, E.;Monge, A. J. Chem. Soc., Dalton Trans. 1998, 1299−1305.

Inorganic Chemistry Article

dx.doi.org/10.1021/ic4025102 | Inorg. Chem. 2014, 53, 1039−10471046

(20) Bryan, J. C.; Mayer, J. M. J. Am. Chem. Soc. 1990, 112, 2298−2308.(21) Hall, K. A.; Mayer, J. M. J. Am. Chem. Soc. 1992, 114, 10402−10411.(22) SMART Data Collection and SAINT-Plus Data ProcessingSoftware for the SMART System (various versions); Bruker AnalyticalX-ray Instruments, Inc.: Madison, WI, 2000.(23) SADABS; Bruker AXS Inc.: Madison, WI, 2001.(24) Sheldrick, G. M. SHELXTL, version 5.1; Bruker AXS: Madison,WI, 1998.(25) Sheldrick, G. M. SHELXL-97; University of Gottingen:Gottingen, Germany, 1997.(26) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci,B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H.P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.;Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima,T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, J. A., Jr.;Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin,K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.;Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega,N.; Millam, J. M.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.;Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.;Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.;Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.;Dannenberg, J. J.; Dapprich, S. Daniels, A. D.; Farkas, O.; Foresman,J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian 09, revision A.02;Gaussian, Inc.: Wallingford, CT, 2009.(27) Huzinaga, S.; Andzelm, J.; Klobukowski, M.; Radzio-Andzelm,E.; Sakai, Y.; Tatewaki, H. Gaussian Basis Sets for Molecular OrbitalCalculations; Elsevier: Amsterdam, The Netherlands, 1984.(28) The stereochemistry of the complexes under study will bedescribed by the configuration index according to the Cahn−Ingold−Prelog system in an octahedral complex OC-6-x-y with the prioritysequence used as follows: O > Npyrrolato > Nimine > Cisocyanide. The firstindex x refers to the ligand priority of the ligand trans to the ligand ofthe highest priority (axial ligands) and the second index y refers to theligand priority trans to that ligand of the equatorial plane, which hasthe highest priority of these four equatorial ligands. Thus,molybdenum(IV) complex 1 possesses OC-6-2-3 stereochemistry, asdepicted in Chart 1.(29) Cahn, R. S.; Ingold, C. K.; Prelog, V. Angew. Chem., Int. Ed. Engl.1966, 5, 385−415.(30) Prelog, V.; Helmchen, G. Angew. Chem., Int. Ed. Engl. 1982, 21,567−583.(31) Block, B. P.; Powell, W. H.; Fernelius, W. C. Inorganic chemicalnomenclature:principles and practice; ACS Professional Reference Book;American Chemical Society: Washington, DC, 1990.(32) von Zelewsky, A. Stereochemistry of Coordination Compounds;John Wiley & Sons Ltd.: New York, 1996.(33) Koch, W.; Holthausen, M. C. A Chemist’s Guide to DensityFunctional Theory; Wiley-VCH: Weinheim, Germany, 2001; p 134.(34) Lahiti, D. W.; Espenson, J. H. J. Am. Chem. Soc. 2001, 123,6014−6024.(35) (a) Kail, B.; Nemykin, V. N.; Davie, S. R.; Carrano, C. J.;Hammes, B.; Basu, P. Inorg. Chem. 2002, 41, 1281−1291. (b) Hoffman,J. T.; Einwaechter, S.; Chohan, B. S.; Basu, P.; Carrano, C. J. Inorg.Chem. 2004, 43, 7573−7575. (c) Hoffman, J. T.; Tran, B. L.; Carrano,C. J. Dalton Trans. 2006, 3822−3830. (d) Kail, B. W.; Basu, P. DaltonTrans. 2006, 1419−1423.(36) Jung, J.-H.; Albright, T. A.; Hoffman, D. M.; Lee, T. R. J. Chem.Soc., Dalton Trans. 1999, 4487−4494.(37) Yamamoto, T.; Yamamoto, A.; Ikeda, S. Bull. Chem. Soc. Jpn.1972, 45, 1104−1110.(38) Helm, L.; Merbach, A. E. Chem. Rev. 2005, 105, 1923−1959.

Inorganic Chemistry Article

dx.doi.org/10.1021/ic4025102 | Inorg. Chem. 2014, 53, 1039−10471047